Click here to watch the video on YouTube!

In this month's episode of Ask a Human: Answers to Unanswered Questions, Attorneys Kassi Burns (King & Spalding, LLP) and Trish McAllister (State Bar of Texas, Law Practice Management Department) answer questions about confidentiality in the age of AI.

If you have any more questions, feel free to comment below! You can also send your questions to Trish McAllister at trish.mcallister@texasbar.com.

Questions:

• How safe is it to use AI? How can I use client information and ensure it remains confidential?
• Can we really ever ensure 100 percent confidentiality and security when using AI for our clients? Is an encrypted question-and-answer AI program required?
• Do closed systems have to be completely local to use confidential information without risking a breach? Are there any web-based utilities that can guarantee secure treatment of confidential material?
• ChatGPT allows a user to deactivate the learning component of a user's chat. Is that sufficient to protect client information if I were to include it in a chat?
• Many software packages using AI to help attorneys write or evaluate contracts require the user to upload the contract to AI software, but that can violate the proprietary nature of the contract. So using AI is limited by what other obligations exist, correct?
• How can I use AI to help me review and pinpoint key items in discovery without breaking confidentiality and ethical rules?
• What practical steps do you suggest that an in-house counsel implement to protect their company data from being harvested by AI for others to use?

Click here to access the AI toolkit.